Artificial intelligence is moving beyond looking at individual cancer genes to understanding the entire tissue landscape, helping pharmaceutical companies develop smarter cancer drugs faster. A major collaboration between Nucleai, an Israeli AI company specializing in tissue analysis, and Gilead Sciences demonstrates how AI-powered tissue intelligence is reshaping biomarker discovery and translational research in oncology drug development.

Why Is Understanding Tissue Architecture Becoming Critical for Cancer Drug Development?

For decades, cancer researchers focused on identifying specific molecular mutations or protein targets. But modern cancer drugs, particularly antibody-drug conjugates (ADCs), which are engineered molecules that deliver toxic payloads directly to cancer cells, require a deeper understanding of how tumors are organized. The tissue's spatial context, the arrangement of different cell types, and the tumor microenvironment all influence whether a drug will work.

Nucleai's AI platform transforms routine pathology images, including hematoxylin and eosin (H&E) stains and immunohistochemistry (IHC) images, into quantitative biological insights. Rather than relying on human pathologists to manually examine slides, the AI analyzes thousands of whole-slide images and connects tissue patterns directly to patient outcomes. In the Gilead collaboration, Nucleai analyzed large datasets of tissue samples across multiple clinical studies spanning several oncology indications, generating novel biological insights and candidate spatial biomarkers.

"Precision oncology is entering a new phase, where understanding tissue architecture is becoming just as important as understanding molecular alterations," said Avi Veidman, Chief Executive Officer of Nucleai.

Avi Veidman, Chief Executive Officer of Nucleai

How Does AI-Powered Tissue Intelligence Differ From Traditional Pathology Analysis?

Traditional image analysis approaches rely on pathologists to manually assess slides, a process that is time-consuming, subjective, and difficult to scale across thousands of clinical samples. Nucleai's AI-native platform integrates computational pathology, spatial biology, clinical outcomes, and multimodal data into a unified framework. This enables standardized biomarker assessment from preclinical research through late-stage clinical development while revealing mechanisms of response, resistance, and disease progression.

The key advantage is reproducibility and scale. When pharmaceutical companies develop companion diagnostics, tests that predict which patients will benefit from a specific drug, they need consistent, objective measurements across thousands of samples. AI achieves this by applying the same analytical rules to every image, eliminating human variability.

Steps to Implement AI-Powered Tissue Analysis in Drug Development

• Digitize Pathology Archives: Convert existing tissue slides and pathology images into digital formats that AI systems can analyze at scale, creating a foundation for retrospective biomarker discovery across past clinical trials.
• Integrate Clinical Outcomes Data: Link tissue analysis results directly to patient outcomes, treatment responses, and disease progression data so the AI learns which tissue patterns predict therapeutic success or failure.
• Validate Spatial Biomarkers: Use AI-identified tissue patterns as candidate biomarkers in prospective clinical studies to confirm their predictive value before incorporating them into companion diagnostic strategies.
• Standardize Across Studies: Apply the same AI analysis framework across multiple clinical trials and indications to ensure consistent, reproducible biomarker assessments that regulatory agencies can trust.

What Makes This Collaboration Significant for the Broader Pharma Industry?

The Gilead-Nucleai partnership signals a broader industry shift. As target expression alone proves insufficient to explain ADC response, pharmaceutical companies are recognizing that protein expression must be understood within the context of tissue architecture, tumor heterogeneity, and microenvironmental factors. This is particularly important for ADCs, which require precise understanding of where target proteins are located and how accessible they are to the drug.

"Scale and reproducibility are becoming essential requirements for biomarker development," explained Dr. Ken Bloom, Head of Pathology at Nucleai.

Dr. Ken Bloom, Head of Pathology at Nucleai

The collaboration demonstrates how AI-powered tissue intelligence can standardize biomarker analyses while connecting tissue biology to clinical outcomes, creating a scalable foundation for translational research, biomarker development, and precision medicine. Nucleai continues to expand collaborations with leading pharmaceutical companies to advance biomarker discovery, translational medicine, companion diagnostic development, and AI-powered tissue intelligence across the oncology development lifecycle.

For drug developers, the practical implication is clear: AI-driven tissue analysis accelerates the path from preclinical research to clinical evidence. By identifying which patients are most likely to benefit from a drug before large-scale trials begin, companies can design more efficient studies, reduce development timelines, and bring effective treatments to patients faster. This represents a fundamental shift in how precision oncology is being developed and validated.